Equilibria in Solution. I. Ion Solvation and Mixed Solvent Interaction

Donald J. Glover. J. Am. Chem. Soc. , 1965, 87 (23), pp 5275–5279. DOI: 10.1021/ja00951a002. Publication Date: December 1965. ACS Legacy Archive...
0 downloads 0 Views 588KB Size
room temperature, the amount of the gas in J at any time was easily determined. The difference between the initial amount of the gas in J and the amount at any given time represents the amount of the gas present in G. After recording pressures on E and F and the room temperature, the tube G was maintained at Dry Ice temperature, while the gas in J was condensed with a liquid nitrogen bath. The mercury in valve H was lowered sufficiently so that a small amount of hydrogen halide in G would bubble slowly intQ J. The coarse sintered glass disk in the right-hand side of valve H prevented mercury from being carried over into J. The valve H was then closed and the constant temperature bath once again raised around tube G. The gas in J was allowed to warm to room temperature, and, after equilibrium had been established in tube G, pressures and room temperature were again recorded. Additional quantities of hydrogen halide were removed from the solubility tube in the manner just described until a sufficient number of points (usually about four) had been taken so that a plot of pressure of hydrogen halide in G vs. mole fraction of the hydrogen halide in solution could be made. Data for a typical determina-

Table VI. Solubility of Hydrogen Chloride in 0.5 M n-Heptane Solution of Benzene at -78.5"

mm.

HC1 in soh, moles

94.00 85.65 71.40 53.45 34.60

1.151 1.046 0.869 0.641 0.410

PHClr

NHCI X

103" 29.59 26.97 22.51 16.70 10.75

Henry's law

const., m. 3180 3180 3170 3200 3220 Av. 3190

-

Benzene = 2.413 mmoles; n-heptane = 35.31 m o l e s .

tion of the solubility of hydrogen chloride in 0.5 M nheptane solution of benzene at -78.5" are summarized in Table VI. The following materials were used for the constant temperature baths: Dry Ice at sublimation pressure of 760 mm., -78.5'; chloroform solid-liquid slush, -63.5 O ; and chlorobenzene solid-liquid slush, -45.2'. The apparatus used to maintain Dry Ice at sublimation pressure of 760 mm. was adequately described previously. lo

Equilibria in Solution. I. Ion Solvation and Mixed Solvent Interaction Donald J. Glover Contribution f r o m the U.S. Naval Ordnance Laboratory, White Oak, Silver Spring, Maryland. Received August 2,1965 A maximum in the p k values of 2,2',4,4',6,6'-hexanitrodiphenylamine (HND)is observed in the system dioxanewater. The maximum is attributed to the solvation of hydrogen ion by water or dioxane in excess of a dioxanewater complex. The complex consists of two molecules of dioxane and one of water. The fact that there is a decrease in k, of water of about 7 X I O 3 in going f r o m water to 7 0 x dioxane, while the k of HND decreases b y only two- to threefold, is explained by solvation effects. The anion of HND is not considered to be further solvated than un-ionized HND, while the hydroxide ion is thought to be solvated with more than one molecule of solvent. Although the law of mass action is used to express equilibria in solution, no general attempt to incorporate the solvent into the expression has been made. There have been correlations of acid ionization data in various solvents, the most notable being the work of Grunwald and co-workers. Much has been written about the effect of changes in dielectric constant on ionization constants. Usually a plot of pk vs. the reciprocal of the dielectric constant is presented, which is based on the Born equation. (1) (a) H. P. Marshall and E. Grunwald, J. Am. Chem. Sac., 76,2000 (1954); (b) E. Grunwald and B. J. Berkowitz, ibid., 73, 4939 (1951); (c) B.Gutbezahl and E. Grunwald, ibid., 75, 559 (1953).

Invariably, these plots are linear only over a very small range of dielectric constant change. As Harned and Owen2' observed, when the plots are extended over a great range such as represented by going from water to 8 2 x dioxane-water, the linearity fails. Such a plot is improved by considering the addition of a water concentration term.3 Even here the extension of the linear plot is not great. It would seem that a consideration of the direct participation of the solvent in the ionization would lead to a better understanding of the solvent composition and the composition of ion solvates. For example, such a participation might be implied by the finding of Harned and Owenzbthat a plot of pk vs. mole fraction of dioxane gives a nearly linear plot. The marked changes in ionization in the system dioxane-water make this system particularly intriguing. Generally, a neutral proton acid decreases in strength from its value in water as the dioxane concentration is increased to 82 %, a decrease of lo5 or lo6 being usual. It is not unusual for the same decrease to be observed (2) H. S. Harned and B. B. Owen, "The Physical Chemistry of Electrolytic Solutions," 3rd Ed., Reinhold Publishing Corp., New York, N. Y., 1958: (a) p. 682; (b) p. 662; (c) p. 756. (3) M. Yasuda, Bull. Chem. SOC.Japan, 32, 429 (1959), introduces (HOH) into the ionization expression for a weak acid in order to get a better plot of pk vs. the reciprocal of the dielectric constant. Curvature still occurs at high concentrations of organic solvent.

Glover 1 Zon Solvation and Mixed Solvent Interaction

5275

of the Potassium Salt of HND

Table I. Molar Absorbancy Indexes ( X Solvent

340

Water Wt. dioxane in water 40.2 79.4 87.4 89.2 90.7 91.1 100.0 Wt. % methanol in water 31.0 86.6 99.8 Wt. % acetone in water 42.0 76.5 100.0

z’

360

370

380

390

400

410

Wave length, mp 420 430 440 460

500

510 520 530 540 550 560

10.8 14.6 17.2 19.6 21.5 23.0 25.1 25.8 26.0 25.4 22.0 16.7 10.7 . . . 5 . 7

2.4

... 1.0

10.9 14.6 17.3 10.1 14.0 17.0 . . . 14.9 17.9 9.9 13.9 16.8 9.9 13.7 16.5 10.4 14.2 16.9 9 . 9 13.4 15.6

2.2 1.5 1.6 1.4 1.7 1.8 2.4

1.4 0.8 1.0 0.8 1.0 1.2 1.6

20.1 20.1 21.0 19.9 19.3 19.7 17.3

22.7 23.1 24.1 22.9 22.1 22.2 18.4

25.2 25.8 21.0 25.6 24.7 24.5 19.0

27.2 28.0 29.1 27.6 26.7 26.1 19.3

28.1 29.1 30.0 28.3 27.4 26.9 18.8

28.1 28.9 29.8 28.0 27.1 26.3 18.2

27.0 27.5 28.4 27.0 25.9 25.1 17.4

23.4 23.3 23.9 22.9 21.9 21.0 15.3

17.5 11.0 8.0 5.3 3.4 17.0 10.0 6.9 4 . 3 2.6 17.4 10.4 7 . 2 4 . 5 2.8 16.7 9 . 9 6.9 4 . 2 2.6 16.0 9 . 7 6 . 8 4 . 5 2 . 8 15.5 9 . 6 6 . 9 4.6 2 . 9 12.5 8.8 6 . 8 4.9 3 . 5

0.7 0.4 0.5 0.4 0.6 0.5 0.9

11.0 15.2 17.9 20.4 23.0 25.0 26.5 27.2 26.9 25.5 21.7 16.1 10.2 7 . 6 5 . 1 3.7 . . . , . . . . . 11.8 16.7 19.7 22.7 25.4 28.0 29.6 29.6 28.8 26.9 21.8 15.2 8 . 4 . . . 3.6 , . . 1 . 4 , . . 0 . 5 11.2 16.5 19.6 22.5 25.4 27.7 29.4 29.5 28.1 26.3 21.0 13.8 7.68 . . . 3.3 . . . 1 . 3 . . . 0 . 4

10.8 1 5 . 1 17.9 20.7 23.4 25.9 27.7 28.3 27.8 26.5 21.9 15.6 12.5 16.3 1 9 . 4 22.2 25.3 27.8 29.8 30.1 29.4 27.7 22.4 15.6 11.6 16.5 19.5 22.7 25.3 28.0 29.7 29.7 28.8 26.8 21.5 14.7

over the total concentration range in ethanol or methanol. There are much data in the literature for ionizations in solutions having concentrations up to 82z by weight dioxane. The search for a suitable neutral proton acid showed a scarcity of ionization data for the total composition range in the system dioxanewater. Kertes and Goldschmidt4 report that the pk of 2,2 ’,4,4’,6,6’-hexanitrodiphenylamine(HND) in dioxane is 0.35, while Schill and Danielssons report a value of 2.81 in water. Apparently, the value of 0.35 is in fact the pH meter reading in dioxane-ethanol at half-neutralization. However, these values indicated that this acid might allow the total dioxane-water composition to be studied. There are other values of the pk of HND reported for but they appear to be in error. In this laboratory, the pk of HND was determined in water, in water-organic solvent mixtures, and in pure organic solvents spectrophotometrically at 28 =t 1’. The ionization is expressed by eq. 1 and 2,where HA is the un-ionized acid, A- is the anion, and the parentheses denotes concentration. kl

HA

H+ + A-

Experimental Section

Apparatus. All spectra were determined on a Beckman spectrophotometer, Model DU, using 1-cm. quartz cells unless stated otherwise. All data were obtained at 28 =t I ” , which was the temperature in the cell compartment. (4) S. Kertes and J. M. E. Goldschmidt, J. Chem. SOC.,401 (195Q ( 5 ) G. Schill and B. Danielsson, Anal. Chim.Acta, 21, 248 (1959). (6) W. D. Treadwell and H. Hepenstrick, Helv. Chim.Acta, 32, 1903 (1949). (7) I